# dtyq/magic

Magicrew. The first open-source all-in-one AI productivity platform (Generalist AI Agent + Workflow Engine + IM + Online collaborative office system)

Repository: https://github.com/dtyq/magic
Canonical: https://ross.abutalabs.com/products/dtyq-magic
Homepage: https://www.magicrew.ai/
Language: TypeScript
License: NOASSERTION
License Family: other
Topics: agent, agi, ai, gpt, llm, low-code, mcp, no-code, sandbox, workflow, super-magic
Last push: 2026-08-12T12:46:11+00:00

## Health v2 (maintenance only)
Score: 63/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 97, release rhythm 35, longevity 34
- inputs: {"age_days": 476, "days_push": 21, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4994, forks 555 (observed 2026-08-28T04:09:04.373862+00:00)

## What it is
MagiCrew (dtyq/magic) is an open-source, enterprise-grade all-in-one AI productivity platform combining a generalist AI agent, workflow engine, instant messaging, and online collaborative office system. It lets organizations deploy AI 'digital workers' with budget controls, human approval gates, sandbox isolation, and deliverable outputs like reports and dashboards.

## Use cases
- deploy enterprise AI agents for real business tasks
- build low-code AI workflows for my team
- self-host an AI agent platform with budget controls
- encapsulate ERP/CRM knowledge into reusable AI workers
- orchestrate multiple AI agents working in parallel
- add human approval gates to risky AI actions
- track AI API costs by department and user
- turn AI output into PPTs, reports, and dashboards

## When to choose
- you need an enterprise AI agent platform with governance, budget caps, and approval workflows
- you want to self-host AI automation with sandbox isolation so data stays in-house
- you want agents that produce actual deliverables (PPT, Excel, dashboards) rather than plain text
- you need multi-agent collaboration plus IM and collaborative office in one platform

## When to avoid
- you only need a lightweight personal AI chatbot or coding assistant
- you want a simple single-purpose agent framework library to embed in your own code
- you require a fully permissive open-source license (license is non-standard)
- you need a mature, battle-tested product with a long enterprise track record

## Facets
- artifact type: application
- maturity: active
- function: agent-framework, workflow-automation, chatbot, mcp, chat-interface, llm-inference, self-hosted
- domain: large-language-models, developer-tools, erp, self-hosted, web-development
- platform: self-hosted, cross-platform
- tags: ai-agents, workflow-engine, digital-workers, enterprise-ai, low-code, no-code, im, collaborative-office, sandbox-isolation, budget-control, multi-agent, skill-creator, automation, web-server, docker, typescript

## Member repositories
- dtyq/magic (main) score 63

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:04.373862+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:17:38.033638+00:00, confidence not recorded.
  - readme: https://github.com/dtyq/magic (fetched 2026-08-28T04:09:04.373862+00:00, sha a76c5a028c86)
  - homepage: https://www.magicrew.ai/ (fetched 2026-08-29T08:59:01.468041+00:00, sha 7b5463a90838)
  - site_page: https://www.magicrew.ai/pricing (fetched 2026-08-29T08:59:01.477263+00:00, sha da5dcb9224a0)
- Data as of 2026-08-30T08:39:29.467469+00:00.